| Agenda | Tutorials |
Tutorial D |
AI Aided Design and Development for Space Systems |
Fees |
$250 USD each |
Date |
Monday – February 24, 2025 |
Time |
8:00 AM – 11:30 AM PT |
Overview
|
The cybersecurity posture of the space enterprise continues to evolve as its architecture modernizes and with the rapid commercialization of space solutions. One key aspect of this evolution is an increasing focus on cybersecurity. The complexity of cybersecurity terminology, standards, and policies can lead to ambiguous and incomplete requirements for system design and acquisition. As such, establishing common understanding and language for system-specific tailoring often takes a great deal of time and effort to incorporate the applicable standards and best practices (e.g., external, internal, de jure, de facto). Recent advances in Large Language Model (LLM) technology provide a potential approach for performing this tailoring in an automated fashion. However, cybersecurity ontologies employed in commercially available Large Language Models are limited to general terminology that may not include the conceptual knowledge required to convey the nuances of space enterprise requirements, focusing instead on more general areas including threat management, vulnerability scanning, and incident response. This general knowledge in a jack of all trades approach is insufficient for emerging space designs and security compliance.This effort done jointly with Aerospace Corp and University of Southern California (USC) Information Sciences Institute (ISI), aims to develop a framework to bridge the gap between requirements for cybersecurity and space system design. The framework follows a modular approach to enable iterative development of a pilot incorporating both domain-specific data and future concepts as they arise. The local LLM implementation will be suitable to safeguard the sensitive data. This presentation will share our initial findings.We map the cybersecurity requirements to space system design by retrieval augmented generation along with a wide range of tools for query rewrite, vector embeddings, and re-ranking and generation for relevant, contextual, and accurate responses
|
| Instructors | Robert Lai, The Aerospace Corporation, Dr. Alefiya Hussain, USC-ISI |
Biography |
Robert Lai CISSP-ISSAP, ISSEP, CGRC, CEH, CSSLP is a Cybersecurity Professional with broad knowledge of systems engineering of system-of-systems, cyber security and secure software lifecycle development. He is a frequent item writer for the CGRC, CISSP, CSSLP, ISSAP and ISSEP certification exams. He was one of the contributors who were instrumental in developing the Certified Secure Software Lifecycle Professional (CSSLP) certification. He was a Lead Chinese Intelligence Analyst for the Cyber Security Forum Initiative’s L7DA (Layer 7 DDOS Attack) Report. He was the first chairperson of the scheme committee for EC-Council. Dr. Alefiya Hussain 20+ years of expertise in networked and distributed systems for security and performance including ground and space systems. She has led several DARPA, NSF, DHS projects in artificial intelligence and machine learning, focused on cybersecurity for large and complex distributed systems. The topics include data engineering with analytics, data fusion, digital twin modeling ranging fromsimulation to testbed emulation of large systems. |
Description of Intended Audience and Recommended Prerequisites |
Basic understanding of systems engineering |
What can Attendees Expect to Learn |
|
| Tutorials |